Computational and Applied Mathematics (2022) 41:312 https://doi.org/10.1007/s40314-022-01995-z Prediction of thermal and energy transport of MHD Sutterby hybrid nanofluid flow with activation energy using Group Method of Data Handling (GMDH) S. Gopi Krishna 1 · M. Shanmugapriya 1 · Ammar Alsinai 2 · Abdu Alameri 3 Received: 7 May 2022 / Revised: 10 July 2022 / Accepted: 6 August 2022 © The Author(s) under exclusive licence to Sociedade Brasileira de Matemática Aplicada e Computacional 2022 Abstract The present research work pursues GMDH for predicting thermal and energy transport of 2-D radiative magnetohydrodynamic (MHD) flow of hybrid Sutterby nanofluid across a moving wedge with activation energy. An exclusive class of nanoparticles SWCNT–Fe 3 O 4 and MWCNT–Fe 3 O 4 are dispersed into the ethylene glycol as regular fluid. The hybrid nanofluid mathematical model has been written as a system of partial differential equa- tions (PDEs), which are then converted into ordinary differential equations (ODEs) through similarity replacements. Numerical solutions are attained Runge–Kutta–Fehlberg’s fourth fifth-order (RKF-45) scheme by adopting the shooting technique. The ranges of diverse sundry parameters used in our study are Hartree parameter 0.1 ≤ m ≤ 0.5, magnetic parameter 0.3 ≤ M ≤ 1, Deborah number 0.1 ≤ De ≤ 1, moving wedge parameter 0.3 ≤ γ ≤ 0.9, Reynolds number 0 ≤ Re ≤ 2.5, solid volume fraction of Fe 3 O 4 and CNTs 0.005 ≤ ϕ 1 ≤ 0.1,0.005 ≤ ϕ 2 ≤ 0.06, Browanian motion 0.1 ≤ Nb ≤ 0.4, ther- mophoresis parameter 0.1 ≤ Nt ≤ 0.25, Eckeret number 0.05 ≤ Ec ≤ 1, radiation parameter 1 ≤ R d ≤ 2.5, Lewis number 0.5 ≤ Le ≤ 1.5, chemical reaction rate 0.1 ≤ σ ≤ 0.7, heat source parameter, 0 ≤ δ ≤ 1.5 and activation energy 1 ≤ E ≤ 4 which shows up during the speed, thermal, and focus for Fe 3 O 4 /C 2 H 6 O 2 nanofluid and CNTs−Fe 3 O 4 /C 2 H 6 O 2 hybrid nanofluid. Additionally, the friction coefficient ( ⌣ C fx ), rate of heat transport ( ⌣ H tx ), and rate of nanoparticle transport ( ⌣ Nt x are calculated using GMDH. The numerical results for the current analysis are illustrated via tables, graphs, and contour plots. The efficiency of the proposed GMDH models is assessed using statistical measures such as MSE, MAE, RMSE, R, Error mean and Error StD. The predicted values are very close to the numerical results, Communicated by Zdenko Takac. B Abdu Alameri a.alameri2222@gmail.com 1 Department of Mathematics, Sri Sivasubramaniya Nadar College of Engineering, Kalavakkam 603110, India 2 Department of Mathematics, University of Mysore, Mysore, Karnataka, India 3 Department of Biomedical Engineering, Faculty of Engineering, University of Science and Technology, Taiz, Yemen 0123456789().: V,-vol 123